Loading...
Please wait, while we are loading the content...
Similar Documents
Automatic User Interaction Correction via Multi-label Graph Cuts
| Content Provider | CiteSeerX |
|---|---|
| Author | Primo, Carlos Escalera, Sergio |
| Abstract | Most applications in image segmentation requires from user interaction in order to achieve accurate results. How-ever, user wants to achieve the desired segmentation accu-racy reducing effort of manual labelling. In this work, we extend standard multi-label α-expansion Graph Cut algo-rithm so that it analyzes the interaction of the user in order to modify the object model and improve final segmentation of objects. The approach is inspired in the fact that fast user interactions may introduce some pixel errors confusing ob-ject and background. Our results with different degrees of user interaction and input errors show high performance of the proposed approach on a multi-label human limb seg-mentation problem compared with classical α-expansion algorithm. 1. |
| File Format | |
| Access Restriction | Open |
| Subject Keyword | Multi-label Graph Cut User Interaction Automatic User Interaction Correction Input Error Accurate Result Multi-label Human Limb Seg-mentation Problem Classical Expansion Algorithm Desired Segmentation Accu-racy Image Segmentation Manual Labelling Object Model Pixel Error Final Segmentation Different Degree High Performance |
| Content Type | Text |
